englishforitprofessionalsfandomcom-20200214-history
Talk:Talking point:Prospects for the Development of Electronics/@comment-128.72.4.146-20130419174732
An atomic force microscope nanoscalpel for nanolithography and biological applications Link: [Nanoscalpel] …'4. Nanolithography applications of the nanoscalpel' We found that the nanoscalpel is an excellent tool for making trenches of different depths in gold layers. Using it with an AFM enabled us to control both the force applied to the scalpel and the depth of the cut. The inset in ﬁgure 4 shows three trenches made in a gold layer on a silicon wafer coated with SiO2. One trench has a depth of ~25 nm, while the other two have depths of ~45 nm and so completely penetrate the metal ﬁlm. In each case the process started with moving the scalpel to a starting position which was usually ~200 nm from the edge of the gold layer. The scalpel was then moved horizontally towards the gold layer with a velocity of 5–10 nm s^(−1). Initially (for X between 0 and 230 nm) the nanoscalpel slides along the comparatively hard SiO2 surface without cutting it. At X = 230 one of the edges of the nanoscalpel comes into contact with the gold layer. The steep rise of the AFM tip in the range of X between 230 and 420 nm corresponds to the process of passing of the nanoscalpel through the gold ﬁlm edge. Then, for X > 420 nm, the deﬂection curve saturates as the scalpel moves along the surface of the gold ﬁlm cutting a 4 nm deep trench in it. The deﬂection in this saturated region is very uniform, indicating that the depth of the cut is fairly uniform (~4 nm), with small variations of 1–2 nm, mostly due to the roughness of the gold surface. This roughness is similar to that observed after scratching metal ﬁlms using unmodiﬁed AFM probes 26. Figure 5 shows how the depth of cuts increases with increasing vertical force applied to the nanoscalpel. In these experiments, the same modiﬁed tip was used ﬁrst for imaging in tapping mode, allowing it to be precisely positioned and then used to cut material at a target site. The applied force was varied by programming different approach depths in the AFM, giving scratches with a range of depths. The sample was then imaged using an unmodiﬁed AFM probe to provide a good spatial resolution and an accurate measure of the scratch depth. For larger forces there was an abrupt increase in depth to ~20 nm as the layer was penetrated completely. AFM measurements showed that no material remained inside the cut, indicating that the metal layer was removed completely (including the titanium adhesion layer). The inset in ﬁgure 5 shows a proﬁle across one of the trenches measured using an unmodiﬁed AFM tip; this gives an accurate measure of the cut depth, although the shape of the cut is not accurately measured due to convolution with the shape of the AFM tip. The edges of the trench are probably in reality more vertical as the cross-section should be similar to the proﬁle of the nanoscalpel. Since the nanoscalpel is capable of completely cutting through deposited gold layers it follows that mechanical lithography on gold ﬁlms should be possible. Currently structures are commonly fabricated by electron beam lithography capable of routinely achieving 50 nm resolution. However, due to drift in electron optical systems, straight line features are difﬁcult to create with such precision, and complicated geometries are limited by electron scattering within resist layers, causing local variations in exposure and a low yield at high resolution. We therefore investigated the use of the nanoscalpel for forming different electrode conﬁgurations by ﬁrst depositing gold structures on an SiO2 surface and then scratching with the nanoscalpel. This process is cleaner than resist-based lithography since no photopolymer is required. The shape of the cut also corresponds to the shape of the scalpel and so is highly reproducible. Figure 6 shows some examples of electrode conﬁgurations fabricated using the nanoscalpel. Figures 6(a) and (b) show 20 nm-wide cuts made across two bridges (20 nm thick gold layer). The image in ﬁgure 6(a) was obtained with the same nanoscalpel that was used for making the cut. While the image is distorted parallel to the plane of the nanoscalpel blade due to the shape of the modiﬁed tip, it can still be used to align the nanoscalpel with the targeted cutting site. Figure 6(b) is a pseudo-3D image captured using a sharp unmodiﬁed tip, which demonstrates better spatial resolution and shows the successful fabrication of the contacts. The material removed from the cut by the scalpel is observed to be deposited on one side of the cut. Figure 6© shows a four-electrode junction fabricated using a similar technique, using two perpendicular cuts in a cross-shaped gold structure. Figure 6(d) shows the formation of a triangular gap by the removal of a segment of the gold bridge by manipulation with the AFM tip after cutting with the nanoscalpel. These experiments demonstrate a number of advantages of the nanoscalpel design for nanolithography. The nanoscalpel is robust; we were able to cut dozens trenches into a gold layer without any observable damage to the scalpel blade.Cut widths of 20 nm, for a cut depth of ~20 nm, were achievable; thinner cuts would require thinner blades, which probably could be fabricated by sharpening 20 nm blades using oxygen plasma etching (a method used in 15 to sharpen AFM tips produced using EBID). The shape of the cut is dependent on the nanoscalpel shape and so the scalpel blade should produce a rectangular cut, which is more useful for precise nanofabrication than the triangular cuts produced by unmodiﬁed tips. There are, however, a few disadvantages to the nanoscalpel. It is quite fragile in the direction perpendicular to the scalpel plane so, at present, it is limited to cutting in one direction; in order to change the cut direction the sample itself must be rotated. There is also some deposition of material at the sides of the cut, which can be undesirable for some applications requiring precise geometries. However, we have demonstrated that the nanoscalpel can be used for precise, clean nanolithography producing nanoscale contacts with small gaps which are difﬁcult to produce using standard mask-based nanolithography techniques. … 6. Summary We have successfully fabricated blade-like structures (‘nanoscalpels’) on AFM probes using electron beam induced deposition of amorphous carbon. Nanoscalpels represent durable and versatile tools for nanofabrication and biological nanosurgery and demonstrate many advantages over current nanomanipulation techniques. We have demonstrated that they can be applied for AFM-based nanolithography of deposited metal ﬁlms producing nanoscale contacts which could form the basis for novel electronic devices. As biological tools, we have shown that nanoscalpels can be used to make incisions on the surfaces of cells which could potentially be used for a wide variety of applications in cell biology, including the ‘nanodissection’ of biological objects to expose their internal structures for in situ imaging and investigation. These experiments show that the nanoscalpel is likely to have a wide range of applications in the creation of nanoscale devices and in biotechnology. Questions (by Fadeev I.V. EKT-57M) 1. What method do authors propose as a possible way to make nanoscalpels thinner than 20 nm? 2. Why do authors use the second (unmodified) tip for making images of trenches? 3. What are the disadvantages of nanoscalpel and AFM-lithography techniques? 4. What factors influence the minimal resolution, that can be achieved by electron beam lithography? 5. What other applications (except fabrication of electrodes) of nanoscalpel can you name?